IndisputableMonolith.Gravity.LatticeRicci
IndisputableMonolith/Gravity/LatticeRicci.lean · 155 lines · 11 declarations
show as:
view math explainer →
1import Mathlib
2import IndisputableMonolith.Constants
3import IndisputableMonolith.Gravity.DiscreteCurvature
4
5/-!
6# Step 4: Lattice Ricci Scalar Convergence (Linearized Case)
7
8Defines the lattice Ricci scalar from deficit angles and proves its
9convergence to the continuum Ricci scalar in the linearized (weak-field)
10regime.
11
12## The Key Result
13
14In linearized gravity with metric g = eta + h:
15 R_continuum = -nabla^2(trace h) + div div h
16
17In harmonic gauge (div h_bar = 0):
18 R = -nabla^2(trace h)
19
20The lattice Ricci scalar is defined as the sum of deficit angles
21divided by dual areas. From DiscreteCurvature:
22 curvature_from_deficit(d2h, a) = -d2h
23
24Summing over 3 axes:
25 R_lattice = -(d2h_xx + d2h_yy + d2h_zz) = -nabla^2(trace h)
26
27This IS the continuum Ricci scalar (in harmonic gauge).
28The convergence error is O(a^2) from the lattice Laplacian convergence.
29
30## Scope
31
32This module proves the LINEARIZED case only. The full nonlinear Regge
33convergence (deficit angle -> Riemann tensor -> Ricci) requires
34significantly more machinery and is stated as a hypothesis for the
35nonlinear regime.
36-/
37
38namespace IndisputableMonolith
39namespace Gravity
40namespace LatticeRicci
41
42open Constants DiscreteCurvature
43
44noncomputable section
45
46/-! ## Linearized Ricci Scalar -/
47
48/-- The linearized Ricci scalar from a metric perturbation h.
49 In harmonic gauge: R = -nabla^2(trace h).
50 We represent this via the 3 second derivatives of the trace. -/
51noncomputable def continuum_ricci_linearized (d2h_xx d2h_yy d2h_zz : ℝ) : ℝ :=
52 -(d2h_xx + d2h_yy + d2h_zz)
53
54/-- The lattice Ricci scalar: sum of deficit-angle curvatures over 3 axes. -/
55noncomputable def lattice_ricci (d2h_xx d2h_yy d2h_zz a : ℝ) : ℝ :=
56 curvature_from_deficit (deficit_angle_deformed d2h_xx a) (a ^ 2) +
57 curvature_from_deficit (deficit_angle_deformed d2h_yy a) (a ^ 2) +
58 curvature_from_deficit (deficit_angle_deformed d2h_zz a) (a ^ 2)
59
60/-! ## Convergence Theorem -/
61
62/-- **LINEARIZED RICCI CONVERGENCE THEOREM**:
63 The lattice Ricci scalar equals the continuum linearized Ricci scalar
64 EXACTLY (not just in the limit).
65
66 This is because each deficit-angle curvature = -d2h_kk exactly
67 (from DiscreteCurvature.curvature_equals_d2h), and the sum gives
68 -(d2h_xx + d2h_yy + d2h_zz) = R_continuum.
69
70 The O(a^2) error comes from the SAMPLING of h on the lattice
71 (i.e., the finite-difference approximation of d2h), not from the
72 deficit-angle-to-curvature step itself. -/
73theorem lattice_ricci_equals_continuum (d2h_xx d2h_yy d2h_zz a : ℝ) (ha : a ≠ 0) :
74 lattice_ricci d2h_xx d2h_yy d2h_zz a =
75 continuum_ricci_linearized d2h_xx d2h_yy d2h_zz := by
76 unfold lattice_ricci continuum_ricci_linearized
77 rw [curvature_equals_d2h d2h_xx a ha,
78 curvature_equals_d2h d2h_yy a ha,
79 curvature_equals_d2h d2h_zz a ha]
80 ring
81
82/-- The lattice Ricci scalar vanishes for flat space (all d2h = 0). -/
83theorem lattice_ricci_flat (a : ℝ) (ha : a ≠ 0) :
84 lattice_ricci 0 0 0 a = 0 := by
85 rw [lattice_ricci_equals_continuum 0 0 0 a ha]
86 unfold continuum_ricci_linearized; ring
87
88/-- The Ricci scalar is negative for positive Laplacian of trace h
89 (e.g., a mass concentration that makes nabla^2 Phi > 0). -/
90theorem ricci_negative_for_mass (d2h_xx d2h_yy d2h_zz : ℝ)
91 (hpos : 0 < d2h_xx + d2h_yy + d2h_zz) :
92 continuum_ricci_linearized d2h_xx d2h_yy d2h_zz < 0 := by
93 unfold continuum_ricci_linearized; linarith
94
95/-! ## Connection to Poisson Equation -/
96
97/-- In linearized RS gravity, the Poisson equation is:
98 nabla^2 Phi = (kappa/2) * rho
99 where Phi is the Newtonian potential and rho is the defect density.
100
101 The Ricci scalar is R = -2 * nabla^2 Phi (for trace h = -2*Phi).
102 So R = -kappa * rho.
103
104 This connects the lattice Ricci to the stress-energy source. -/
105noncomputable def ricci_from_source (rho : ℝ) : ℝ :=
106 -ZeroParameterGravity.kappa_rs * rho
107
108theorem ricci_source_proportional_to_kappa (rho : ℝ) :
109 ricci_from_source rho = -(8 * phi ^ 5) * rho := by
110 unfold ricci_from_source
111 rw [ZeroParameterGravity.kappa_rs_closed_form]
112
113/-- Positive mass density gives negative Ricci scalar (attractive gravity). -/
114theorem ricci_attractive (rho : ℝ) (hrho : 0 < rho) :
115 ricci_from_source rho < 0 := by
116 unfold ricci_from_source
117 exact neg_neg_of_neg (mul_pos ZeroParameterGravity.kappa_pos hrho) |>.2
118
119/-! ## Nonlinear Regime (Hypothesis) -/
120
121/-- The FULL (nonlinear) Regge convergence: deficit angles on a refined
122 lattice converge to the Riemann curvature tensor, and the summed
123 Ricci scalar converges at O(a^2).
124
125 This is the content of Regge's 1961 theorem (confirmed numerically
126 by Gentle-Miller 2001). We state it as a hypothesis for the
127 nonlinear regime, with the linearized case proved above. -/
128def H_nonlinear_regge_convergence : Prop :=
129 ∀ (R_exact : ℝ) (a : ℝ), 0 < a → a < 1 →
130 ∃ C : ℝ, 0 < C ∧
131 ∀ d2h_xx d2h_yy d2h_zz : ℝ,
132 |lattice_ricci d2h_xx d2h_yy d2h_zz a - R_exact| ≤ C * a ^ 2
133
134/-! ## Certificate -/
135
136structure LatticeRicciCert where
137 linearized_exact : ∀ d2h_xx d2h_yy d2h_zz a : ℝ, a ≠ 0 →
138 lattice_ricci d2h_xx d2h_yy d2h_zz a =
139 continuum_ricci_linearized d2h_xx d2h_yy d2h_zz
140 flat_vanishes : ∀ a : ℝ, a ≠ 0 → lattice_ricci 0 0 0 a = 0
141 attractive : ∀ rho : ℝ, 0 < rho → ricci_from_source rho < 0
142 kappa_coupling : ∀ rho : ℝ, ricci_from_source rho = -(8 * phi ^ 5) * rho
143
144theorem lattice_ricci_cert : LatticeRicciCert where
145 linearized_exact := lattice_ricci_equals_continuum
146 flat_vanishes := lattice_ricci_flat
147 attractive := ricci_attractive
148 kappa_coupling := ricci_source_proportional_to_kappa
149
150end
151
152end LatticeRicci
153end Gravity
154end IndisputableMonolith
155